We are happy to announce 4 papers accepted to the IEEE International Conference on Robotics & Automation (ICRA) 2026. See you all in Vienna, Austria!
UniFField: A Generalizable Unified Neural Feature Field for Visual, Semantic, and Spatial Uncertainties in Any Scene
Christian Maurer*, Snehal Jauhri*, Sophie Lueth, Georgia Chalvatzaki
We present UniFField, a unified uncertainty-aware neural feature field that combines visual, semantic, and geometric features in a single generalizable representation while also predicting uncertainty in each modality. Our approach, which can be applied zero-shot to any new environment, incrementally integrates RGB-D images into our voxel-based feature representation as the robot explores the scene, simultaneously updating uncertainty estimation.
SE(3)-PoseFlow: Estimating 6D Pose Distributions for Uncertainty-Aware Robotic Manipulation
Yufeng Jin, Niklas Funk, Vignesh Prasad, Zechu Li, Mathias Franzius, Jan Peters, Georgia Chalvatzaki
The paper introduces SE(3)-PoseFlow, a probabilistic framework utilizing Flow Matching on the SE(3) manifold to estimate full 6D object pose distributions, effectively managing ambiguities from symmetries and occlusions.
Stein Variational Ergodic Surface Coverage with SE(3) Constraints
Jiayun Li, Yufeng Jin, Sangli Teng, Dejian Gong, Georgia Chalvatzaki
This paper formulates Stein Variational Gradient Descent (SVGD) on the SE(3) manifold and shows how it relates to ergodic surface coverage on general meshable surfaces.
Adaptive Diffusion Constrained Sampling for Bimanual Robot Manipulation
Haolei Tong*, Yuezhe Zhang*, Sophie Lueth, Georgia Chalvatzaki1
In this work, we propose Adaptive Diffusion Constrained Sampling (ADCS), a generative framework that flexibly integrates both equality (e.g., relative and absolute pose constraints) and structured inequality constraints (e.g., proximity to object surfaces) into an energy-based diffusion model.
